cli

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly routes to wget for fetching remote URLs (see SKILL.md tool table and references/common-use-cases.md and references/wget.md which show commands like "wget -qO- https://... | jq"), meaning the agent can ingest and parse arbitrary public web content that could influence subsequent actions.